from directed_graphs.datasets import directed_swiss_roll_uniform, plot_directed_3d
X, flow, labels = directed_swiss_roll_uniform(num_nodes=1000, num_spirals=2.5, radius=1, height=12, xtilt=0, ytilt=0)
plot_directed_3d(X, flow, labels, mask_prob=0.5)

X = torch.tensor(X).float().to(device)
flow = torch.tensor(flow).float().to(device)
X = X.float().to(device)
flow = flow.float().to(device)

BOBO_FET = DiffusionDistanceFlowEmbedder(X, flow, labels = labels, device = device)

BOBO_FET.fit()

  0%|          | 0/100 [00:00<?, ?it/s]

  1%|          | 1/100 [00:05<08:42,  5.28s/it]

  2%|▏         | 2/100 [00:10<08:39,  5.31s/it]

  3%|▎         | 3/100 [00:15<08:33,  5.30s/it]

  4%|▍         | 4/100 [00:21<08:27,  5.29s/it]

  5%|▌         | 5/100 [00:26<08:21,  5.28s/it]

  6%|▌         | 6/100 [00:31<08:21,  5.33s/it]

  7%|▋         | 7/100 [00:37<08:12,  5.30s/it]

  8%|▊         | 8/100 [00:42<08:07,  5.30s/it]

  9%|▉         | 9/100 [00:47<08:04,  5.32s/it]

BOBO_FET.visualize_embedding()

BOBO_FET.training_gif()

BOBO_FET.visualize_loss()

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Input In [9], in <cell line: 1>()
----> 1 BOBO_FET.visualize_loss()

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/directed_graphs/directed_graphs/flow_embedding_training_utils.py:133, in FETrainer.visualize_loss(self, loss_type)
    131 if loss_type == "all":
    132   for key in self.losses.keys():
--> 133     plt.plot(self.losses[key])
    134   plt.legend(self.losses.keys(), loc='upper right')
    135   plt.title("loss")

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/matplotlib/pyplot.py:2769, in plot(scalex, scaley, data, *args, **kwargs)
   2767 @_copy_docstring_and_deprecators(Axes.plot)
   2768 def plot(*args, scalex=True, scaley=True, data=None, **kwargs):
-> 2769     return gca().plot(
   2770         *args, scalex=scalex, scaley=scaley,
   2771         **({"data": data} if data is not None else {}), **kwargs)

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/matplotlib/axes/_axes.py:1632, in Axes.plot(self, scalex, scaley, data, *args, **kwargs)
   1390 """
   1391 Plot y versus x as lines and/or markers.
   1392 
   (...)
   1629 (``'green'``) or hex strings (``'#008000'``).
   1630 """
   1631 kwargs = cbook.normalize_kwargs(kwargs, mlines.Line2D)
-> 1632 lines = [*self._get_lines(*args, data=data, **kwargs)]
   1633 for line in lines:
   1634     self.add_line(line)

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/matplotlib/axes/_base.py:312, in _process_plot_var_args.__call__(self, data, *args, **kwargs)
    310     this += args[0],
    311     args = args[1:]
--> 312 yield from self._plot_args(this, kwargs)

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/matplotlib/axes/_base.py:490, in _process_plot_var_args._plot_args(self, tup, kwargs, return_kwargs)
    488     y = _check_1d(xy[1])
    489 else:
--> 490     x, y = index_of(xy[-1])
    492 if self.axes.xaxis is not None:
    493     self.axes.xaxis.update_units(x)

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/matplotlib/cbook/__init__.py:1614, in index_of(y)
   1612     pass
   1613 try:
-> 1614     y = _check_1d(y)
   1615 except (np.VisibleDeprecationWarning, ValueError):
   1616     # NumPy 1.19 will warn on ragged input, and we can't actually use it.
   1617     pass

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/matplotlib/cbook/__init__.py:1306, in _check_1d(x)
   1304 x = _unpack_to_numpy(x)
   1305 if not hasattr(x, 'shape') or len(x.shape) < 1:
-> 1306     return np.atleast_1d(x)
   1307 else:
   1308     return x

File <__array_function__ internals>:180, in atleast_1d(*args, **kwargs)

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/numpy/core/shape_base.py:65, in atleast_1d(*arys)
     63 res = []
     64 for ary in arys:
---> 65     ary = asanyarray(ary)
     66     if ary.ndim == 0:
     67         result = ary.reshape(1)

File /gpfs/loomis/project/krishnaswamy_smita/kjm76/conda_envs/flowembed/lib/python3.9/site-packages/torch/_tensor.py:757, in Tensor.__array__(self, dtype)
    755     return handle_torch_function(Tensor.__array__, (self,), self, dtype=dtype)
    756 if dtype is None:
--> 757     return self.numpy()
    758 else:
    759     return self.numpy().astype(dtype, copy=False)

TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

BOBO_FET.losses['diffusion']

100

import glob
import ipywidgets as widgets
from PIL import Image
import os
import ipywidgets as widgets
import base64
frames = [Image.open(image) for image in glob.glob(f"visualizations/{BOBO_FET.timestamp}/*.jpg")]
frame_one = frames[0]
frame_one.save(f"{BOBO_FET.title}.gif", format="GIF", append_images=frames,
           save_all=True, duration=300, loop=0)
# display in jupyter notebook
b64 = base64.b64encode(open(f"{BOBO_FET.title}.gif",'rb').read()).decode('ascii')
display(widgets.HTML(f'<img src="data:image/gif;base64,{b64}" />'))

class DiffusionDistanceFlowEmbedder[source]

`class` `DiffusionDistanceFlowEmbedder`[source]